All AI models · LLM benchmarks · Methodology

Alibaba

Qwen3.5 122B-A10B

Qwen vision-language model for visual reasoning, documents, and agent tasks

Model facts

Context window
262K tokens
Maximum output
66K tokens
Cheapest paid input
$0.115 per 1M tokens
Cheapest paid output
$0.917 per 1M tokens
Open weights
yes
Providers
20

Capabilities and modalities

reasoning, tool calling, structured output, attachments, temperature control, open weights, text, image, audio, video.

Local hardware estimate

At 4K context: Q4_K_M 67.8 GiB, Q8_0 126.0 GiB, F16 235.3 GiB working memory.

Parameters
125.1 billion
Architecture
qwen3_5moe
Model license
apache-2.0

Planning estimate adapted from the MIT-licensed whichllm estimator using metadata from Hugging Face; not a vendor minimum requirement.

Artificial Analysis benchmarks

Compare this model on the full LLM benchmark leaderboard.

ConfigurationAAICodingMathOutput tokens/s
Reasoning32.845.7not measured135.4
Non-reasoning28.243.3not measured151.1

Observed price history

  • 2026-08-24: $0.115 input / $0.917 output per 1M tokens

API providers